Chapter 3

The Viking Diaspora

Introducing the Viking Diaspora

The Long, Broad – Continuities in Time and Space

Archaeologists recognise that is characterised by ‘remarkable patterns of continuity which link the distant past to the present’.1 These patterns can be seen, not just in archaeological evidence, but in many cultural practices particularly associated with the Viking Age. The continuities suggest that some evidence for the

Viking Age is best considered in a chronological context broader than even the generous framework of 750-1100. This broader chronological context was outlined in

Chapter 1, where it was argued that it should be extended to c. 1500. Because the

Viking Age is the period when many Scandinavians left Scandinavia, often permanently, it is also important to widen the geographical range, and consider evidence from all the areas touched by Scandinavian settlement in that period. Just as much of the evidence stretches the chronological boundaries of the Viking Age, so there is also much evidence, whether natural, artefactual or linguistic, that stretches those geographical boundaries, and which can only be considered in the context of the larger Viking world. This geographical framework was outlined in Chapter 2, above.

1 Hodder and Hutson 2003, 140.

1

The purpose of this chapter is to show by means of a small number of examples how this long, broad Viking Age works in practice, in connection with various kinds of natural, artefactual and linguistic evidence for the Viking Age. This does not mean that there is no space for the local, the regional and the otherwise particular. There have been some recent studies which have emphasised local variations in cultural practices even within Scandinavia, let alone outside it, and which have therefore argued against any pan-Scandinavian, unifying concept of the

Viking Age, indeed against any overarching concept including the word ‘Viking’.2 It would of course be surprising if there was not variation of many kinds across a long period of time and a very broad and varied geographical range. Yet certain continuities, both chronological and geographical, must also be present if the Viking

Age and its aftermath are justifiably to be termed a diaspora. Some such continuities are outlined here to demonstrate that, even allowing for local regional variation, some aspects of the Viking Age have a greater reach in both space and time.

The term ‘diaspora’ will also be explored more closely, particularly in relation to this question of continuity vs. variation. It will be suggested that ‘diaspora’ is precisely the term that can resolve this paradox. The processes of diaspora counteract the tendency to variety and difference by selecting and emphasising certain cultural features and thereby creating continuity across time and space, and by discovering or even manufacturing other aspects of continuity. This can be seen across the range of natural, artefactual and linguistic evidence, and across both the chronological and geographical range.

2 E.g. Svanberg 2003, I.

2

Natural Evidence

What has been defined as ‘natural evidence’ (see ch. 1) is in many ways the most difficult to incorporate into a diasporic understanding of the Viking Age, which is predominantly concerned with cultural processes best represented by the artefactual and linguistic evidence. However, natural evidence is still extremely useful in understanding the migrations that were the prerequisite of diaspora, as outlined in

Chapter 2. The natural environment is always changing, and those changes which mark significant events in the Viking Age have to be considered in the context of the environmental and climate change that are a constant in , and which are sometimes caused by humans and sometimes not. The previous chapter has shown some of the impacts of Scandinavian settlers on their new environments, both the pristine and the already inhabited. Some of this research has placed the

Scandinavian environmental impact in a longer historical context, such as the destruction of ’s original woodland, a process that seems to have spanned many centuries from the settlement era to the early modern period. Other research is still in the development phase, for example the strontium stable isotope analysis of sheep’s wool which shows promise for the future provenancing of textiles and therefore a better understanding of patterns of both trade and migration.3 But the most obvious example of natural evidence from a much later period that has been used to illuminate Scandinavian activities in the Viking Age is that of population genetics.

Case Study – Genetics

3 Frei et al. 2009.

3

Most genetic studies purporting to give insights into Viking Age migrations are in fact based on inferences from the genetic patterns of current populations, on the basis that a large enough, well-chosen sample will reflect the history of that population in some measure. A classic example is Iceland which had some important advantages for DNA studies as the technology for these burgeoned in the 1990s. Its population is small, enabling the recording of DNA information for the whole population, primarily for the purposes of medical research, but with side benefits for historical research. And the history of that population is well-known, at least in outline, with no significant immigration since its settlement in the Viking Age, so justifying the assumption that the late twentieth-century population was a good proxy for the founding population over a thousand years ago.4 These studies produced the much- touted results which identified the origins of the as being both in , and in Britain and . As already noted in the previous chapter, this result was not unexpected, but the scientists claimed to be able to establish the nature of these origins in more detail. In particular, they claimed to have demonstrated that there was a considerable difference between the ancestry of the founding male population, over two-thirds of which had DNA similar to the present population of

Norway, and the ancestry of the founding female population, two-thirds of which conversely seem to have had their origins in the Celtic parts of the .5

Since those pioneering studies, there is a greater recognition of the problematic nature of such historical DNA studies based on modern populations.

Comparisons of founding populations with the current populations of their supposed

4 Gillham 2011, 12-19. 5 Helgason et al. 2000a; 2000b; 2001.

4 homelands depend on the genetic evidence for those homeland populations, which may have its own problems of quantity and quality (and nowhere is as thoroughly mapped as Iceland). Also, such comparisons do not take into account any changes there might have been in the homeland populations since the Viking Age, not a topic which has been much studied in, for example, Norway. Furthermore, small populations like Iceland (and even more so the Faroes) are particularly susceptible to genetic drift, in which various factors eliminate some genetic lines from the population creating bottlenecks between past and present population structures.

These factors include disease, in particular epidemics, and famine, both of which are known to have made a substantial reduction in the Icelandic population, and emigration, which was considerable in the nineteenth century.6 Geneticists are of course aware of all of these issues, and they use mathematical modelling to get round some of the problems, but more popular presentations often ignore these problems with the evidence and simplify the results.7 The scientific studies are also subject to reinterpretation in what has been called ‘applied genetic history’. The reduction of an individual’s complex genetic history to a matter of ‘Viking descent’ plays into the creation of individual and familial narratives of origin and belonging, which in turn affect the self-selection of those who submit themselves to testing.8

Any discussion of the value of the genetic evidence must take these factors into account.

In larger populations, particularly in , where surnames have been established since the , the obvious connection between the Y-

6 Gillham 2011, 13; Karlsson 2000, 234-8. 7 Thomas 2013. 8 Scully et al. 2013.

5 chromosome and surnames, both being passed down from father to son, has enabled the better-targeted selection of samples.9 In areas with high immigration in modern times, descendants of these more recent immigrants can be excluded through the selection of subjects with surnames attested in the area in late medieval times, giving a population sample chronologically closer to, and therefore more likely to be representative of, the historical population. A study of just such a population sample in the north-west of England demonstrated a substantial proportion of members (in the region of 50%) whose direct male ancestor had a Y-chromosome type most commonly found in Norway, a fact which was then explained by the

Viking Age settlement of the area already suggested by place-names, archaeology and some documentary sources.10

Thus, studies from both Iceland and the north-west of England have shown that modern population genetics can make a contribution to understanding Viking

Age migrations, though the limitations of and constraints on such evidence must always be borne in mind. Also, like all natural evidence, but unlike, on the whole, the artefactual and the linguistic evidence, genetics provides insights which depend on the deployment of modern scientific methods, insights which could not possibly have been available to people in the Viking Age itself. For this reason, the natural evidence is a useful check on the artefactual and linguistic evidence, which was actually produced by people in the past, and vice versa. For example, the genetic study of the north-west of England can only tell us that some males of Norwegian descent, perhaps in considerable numbers, must have passed through the area and

9 King and Jobling 2009. 10 Bowden et al. 2008; King and Jobling 2009, 356.

6 left their sperm behind. But a study of the artefactual and linguistic evidence from the same region is needed to understand the broader context for this and, in particular, whether this injection of sperm took place in a context of Norse speech and cultural practices, which included women of Norwegian origin or descent as well as men, as will be discussed further in Chapter 4.

Artefactual Evidence

One of the more spectacular indications of archaeological continuity over a long period of time are the farm mounds, or rather settlement mounds, since the inhabitants might also have carried out other types of activities such as fishing.

These are found in various places, most notably in , but also in

Orkney, Faroe and Iceland. 11 They are accumulations of settlement debris which build up as the inhabitants of the settlement discard both household and agricultural rubbish, and renew their living accommodation and farm buildings. They indicate stability and a favoured location. Some of these, with modern farm buildings still on top, have a continuous history of habitation going back around 2000 years, although the majority seem to have started to accumulate around the turn of the last millennium. In places like Sanday, , it is possible to see modern farms still operating on top of mounds which have their origins in the Viking Age migrations to the islands. Even without a farm mound, the Viking Age and late Norse site of

Belmont on Unst, , lasted in essentially the same form for a minimum of

11 Bertelsen and Lamb 1993, 545; Arge 2005, 26; Sveinbjarnardóttir 2011, 261; Harrison 2013.

7

400 years, while Quoygrew, on Westray in Orkney, provides a millennium-long sequence of continuous archaeology from the tenth to the twentieth centuries.12

Apart from actual sites, there are other kinds of continuity which relate to form and type rather than specific instances. Buildings can provide interesting evidence across both time and space. The Viking Age rural dwelling, for instance, was based on a fairly standard model which gradually developed over time into a more complex structure and domestic space. But this standard model also had to be adapted to local conditions, particularly in regard to the available building materials, which would vary enormously from the wood-rich regions of the homelands to the generally treeless settlements of the North Atlantic where the buildings were made of stone or turf. Whether built in wood, stone or turf, the basic shape and dimensions of the three-aisled rectilinear houses with curving walls and about 20 metres long are found in the Northern Isles and across the North Atlantic, suggesting the community of ideas informing their construction, and deriving from models in the Scandinavian homelands.13 Sometimes cultural imperatives would override the constrictions of local building materials, as when wooden buildings were imported wholesale into Iceland from Norway, such as the building known as

Auðunarstofa, built for the Norwegian bishop of Hólar in northern Iceland in 1317.14

Building types also reveal cultural connections. Thus, has the standard

Scandinavian-type , as described above, and found across the North

Atlantic, reflecting perhaps the origins of its settlers in Iceland. But it also has a house-type which reflects Scandinavian urban architecture from the eleventh century

12 Larsen 2013c, 215; Barrett 2012a. 13 Barrett 2012b, 13-15; Larsen 2013c, 214-15. 14 Gunnarsson 2004; ÍF XVII, 326-7; LLBH, p. 62.

8 onwards, and which appears to be evidence for direct contacts between Greenland and Norway after the initial settlement, contact not necessarily mediated through

Iceland.15

Not all artefactual evidence shows continuities with the homeland, or across a long period of time. Some artefacts seem to have been developed as a consequence of the Scandinavian migrations, without any obvious link to any homeland, but with links across several of the overseas settlements, and thus have a wide geographical range. For example, small metal bells that have been termed ‘Norse bells’ are found in a range of contexts and in substantial numbers in England, but also in , the Isle of Man, north Wales, Ireland and Iceland. A recent study has described these as ‘a Scandinavian colonial artefact’ on the grounds that, while there are no parallels for them from the Scandinavian homelands, their distribution is clearly related to a Scandinavian presence in those places where they are found.16 They appear to date mainly to the tenth century and their function is uncertain, though the most likely explanation seems to be that they were used as necklace pendants by high-status women, for the purpose of ostentation, possibly with an amuletic function as well.17 Their distribution suggests a fashion which arose through contact between a range of different Scandinavian communities, including the Danelaw and the Irish Sea region, reflecting the tendency of diaspora to create new cultural forms, as will be discussed further below. Both the gender associations of these little bells and their geographical distribution provide interesting insights into the nature

15 Høegsberg 2009, 98, 103-4. 16 Schoenfelder and Richards 2011, 157. 17 Schoenfelder and Richards 2011, 160.

9 of and links between various Scandinavian-origin communities in the west, but during a relatively restricted period of time.

Case Study – Steatite

A material that is particularly characteristic of the Viking Age in some regions, and enables the tracking of migrations throughout the Viking world, but also illustrates connections across the long Viking Age and around the Viking world, is that of steatite, also know as soapstone or kleber.18 Steatite is a useful mineral, found extensively in Norway, but also in Shetland and Greenland. When newly exposed, it is soft and easily carved with either metal or stone tools. It then hardens with use, or on heating, making it suitable for domestic vessels of all kinds, for both cooking and storing food. Other uses are for textile tools such as spindle whorls and loom weights, lamps, fishing weights, or even beads and gaming counters, often recycled from larger vessels.

The were very used to this handy material: instead of pottery they generally made their domestic vessels from steatite, and the quarrying and working of steatite were major industries in Viking Age Norway.19 With the arrival of

Scandinavians in Shetland, there is a noticeable decrease in ceramic pottery in the archaeological record and an increase in the use of steatite, which characterises the earliest Norse phases at in Shetland but also Pool in Orkney.20 The increase in the quantity of finds in the Viking Age, and the archaeological assessment of the earliest finds as coming from Norway, suggest that these items

18 Much of the following is based on Forster and Turner 2009; Ritchie 1984 is a useful introduction. 19 Baug 2011, 311. 20 Hunter et al. 2007, 139, 412-33; Dockrill et al. 2010, 12, 80-81, 297-301.

10 were brought with the incomers.21 At sites like Belmont, in Unst, the steatite finds include not only fragments of imported Norwegian vessels but also a large amount of both worked and unworked steatite and manufacturing waste.22 There are eight steatite outcrops close to the site and it is likely that the quarrying of steatite and manufacture of objects from it played an important part at Belmont, as elsewhere in

Shetland.23 The new Shetlanders built up a profitable trade in the material, to the neighbouring islands of Orkney in particular, but also to the Faroes and the

Hebrides.24

Steatite goods found at markets such as Kaupang in Norway and Hedeby in

Denmark indicate that they were traded widely. Kaupang had intensive manufacture and distribution of steatite vessels, and seems to have been a centre for their export to southern Scandinavia from around 800, while the steatite found at Hedeby is probably of Norwegian origin.25 A fair number of steatite objects of probably

Norwegian origin have also been found at ninth- and tenth-century sites in Russia, where the artefacts are predominantly of a Scandinavian type, raising the question of whether the steatite was traded or brought there by immigrants.26

Later on, Shetlanders made more use of their local resource, but also continued to import Norwegian vessels. Petrological analysis of steatite is still developing, but it is now possible to distinguish examples deriving from Shetland

21 Owen and Lowe 1999, 170-73, 293; Forster 2009, 65; Dockrill et al. 2010, 93-4, 266-72; Batey et al. 2012, 207. 22 Larsen and Dyhrfjeld-Johnsen 2013, 194-204. 23 Larsen 2013a, 205-6. 24 Larsen 2013a, 206. 25 Baug 2011, 318, 332, 335-6. 26 Khvoshchinskaya 2007.

11 from those manufactured in Norway.27 The extensive outcrops of steatite in Shetland provided a useful source for topping up supplies, a development dated to the second half of the tenth or the eleventh century at Pool, and in Orkney, steatite could be imported from Shetland as well as Norway.28 A new type of vessel commonly made of steatite, known as a bakeplate or a bakestone, was manufactured earlier in

Shetland than Norway, and then reimported to Shetland from Norway.29 Bakeplates were used in Norway from the middle of the eleventh up to the seventeenth century.

At the Shetland sites of , Da Biggins and Sandwick, they are found in late

Norse levels, especially from the thirteenth and fourteenth centuries.30 At Da

Biggins, in particular, the large numbers of these vessels, used for making flatbread and imported from Norway, are part of a larger body of evidence which reveals the close contacts at least some of the inhabitants had with Norway from the twelfth century onwards.31 Steatite fragments from St Kilda also come from a late Norse bakeplate, associated with a pottery fragment dated to 1135 ± 170.32 They are also used in Iceland between c. 1100 and 1500.33

Steatite continued to be valued by Norwegians even after the Viking Age and after the introduction of other materials to make domestic vessels, indeed the use of steatite in that country has been called ‘a cultural trait’.34 The cathedral in

Trondheim, the construction of which began in the twelfth century, is built from a number of different kinds of stone, but from 1200 onwards the main material for

27 Forster and Turner 2009, 117. 28 Hunter et al. 2007, 432; Forster 2009, 68; Dockrill et al. 2010, 272-80; Batey et al. 2012, 212-17. 29 Forster 2009, 65-8; Dockrill et al. 2010, 280-83. 30 Owen and Lowe 1999, 293. 31 Weber 1999; Crawford 1999, 247. 32 Emery and Morrison 1995, 41. 33 Sveinbjarnardóttir 2011, 155-6. 34 Baug 2011, 332.

12 both its construction and its decoration is steatite. The reason for this has been linked to local traditions of using steatite for everyday vessels and other objects

(since there was no tradition of building in stone in Norway before the arrival of

Christianity), and certainly some of the same quarries were used for both vessels and architecture. However, it is more likely to represent the development of local expertise in building with stone.35 The earliest phases of the cathedral would have been built by foreign masons, who used the kinds of stone with which they were familiar. By around 1200, local expertise had developed sufficiently to cast off imported ideas of the most appropriate building stone and use that which was abundantly locally available and most suitable for the purpose. This architectural use of steatite is a good example of the Norwegian, and indeed Scandinavian, ability to take up new ideas and adapt them to their own traditions. It is particularly appropriate that the cathedral containing the relics of Norway’s national should be made of Norway’s national stone.

It is always important to consider negative evidence, too, for example where and when steatite is not used despite cultural propensities. In the , the

Scandinavian immigrants continued the local practice of using ceramic vessels rather than importing steatite as they easily could have done. Whether this was because the immigrants were fewer, or took a different attitude to local customs, is not clear.

Yet even here the Scandinavians put their own stamp on the cultural practices they borrowed. Finds from excavations in the Outer Hebrides show both continuity and change in the production and use of pottery in early Viking contexts. Thus, simple handmade pottery continued to be manufactured using local materials as in the pre-

35 Storemyr 2003.

13

Viking period, but the arrival of the new inhabitants is shown in new manufacturing techniques and the size and shape of vessels made, though the dating of these new forms is still uncertain.36

This continuity in the use of pottery suggests some kind of continuity of population, or at least contact between the old and the new populations, in a way less clearly evidenced in the Northern Isles, although there is some evidence for the production and use of coarse pottery in Unst, despite the proximity of a steatite outcrop.37 This differential pattern between the Western and Northern Isles is also evident in other forms of evidence, such as genetic studies of the modern populations of the different island groups. These suggest that the Hebrides were, like Orkney and Shetland, settled by family groups from Norway, but that the proportion of these in the population was smaller than in Orkney and Shetland, leading to the possibility of greater contact with indigenous inhabitants, as evidenced by the pottery.38 While the use of steatite in places like Shetland is undoubtedly bound up with its local availability as well as connections with Norway, in a place like Orkney it seems to have been a cultural choice. The Viking Age settlers of Orkney chose to import (from Norway or Shetland) the material familiar to their culture, rather than learn or adapt more local traditions of pottery, as happened in the Hebrides. The islands of St Kilda, with very few Viking and Norse finds, nevertheless have some associated finds of both pottery and steatite, which

36 Lane 2010. 37 Brown 2013a; 2013b. 38 Goodacre et al. 2005.

14 are dated to 1135 ± 170, reflecting their intermediate cultural position between the

Hebrides and the Northern Isles.39

Linguistic Evidence

The long and broad Viking Age is most easily demonstrated in various forms of linguistic evidence. The runic inscriptions of the Viking Age have their origins in the development of this form of writing in Scandinavia before 150.40 The runic alphabet then continued to be used for some centuries after the Viking Age and even after the introduction of the roman alphabet, not only in all three of the homelands but also in regions settled in the Viking Age, especially Scotland, Iceland and Greenland.

Less long-lived in time, but even more distant in both physical and cultural space, are the Scandinavian runic inscriptions found in Russia, and as far east and south as the Black Sea, Istanbul and Athens.41

Poetry also shows a remarkable continuity between around 300-1500, as evidenced by some early runic inscriptions as well as later manuscripts.42 The basic structures of Scandinavian poetry, especially its metres and diction, are maintained right through the major change from orality to literacy that happened between the late Viking Age and the twelfth century, accompanied by further changes in the cultural and social functions of poetry through this period.43 Although best attested in medieval Iceland and Norway, the geographical range of Scandinavian poetry is

39 Emery and Morrison 1995, 41. 40 Barnes 2012a, 9-14. 41 Jansson 1987, 61-2; Barnes 2012a, 89. 42 Jesch 2008a. 43 Gade 2000.

15 also wide, including and , and the diasporic lands of England,

Scotland and Greenland, as will be discussed further in Chapter 6.

As already noted (ch. 1), many of the written sources for the Viking Age were composed or put down on vellum, or both, in a later period and it is not possible to study the Viking Age without taking those sources into account. It is not, however, simply a matter of distinguishing between ‘contemporary’ and ‘later’ sources and assuming that the former are preferable. Traditionally, sources such as the Anglo-

Saxon Chronicle or the Frankish annals have been highly valued for their contemporary insights into some aspects of the Viking Age. But such texts usually represent particular political or cultural agendas, and the doings of form only a part of what they choose to record, which is then usually also presented in a way that accords with those agendas.44 Moreover, some such annals which are considered to be contemporary are nevertheless preserved only or mainly in later manuscripts, including most Irish annals.45 Other annalistic texts were clearly written at a later date but seem nevertheless to incorporate some contemporary information, such as the Russian Primary Chronicle.46 The process of historiography is generally an ongoing one, and Viking Age history is indebted to many sources from the twelfth century and later, such as the Irish Cocad Gaedel re Gallaib.47

Similarly, and as already noted (ch. 1), some texts in the Scandinavian vernacular, notably skaldic poetry, are arguably contemporary (oral) sources from the Viking

Age, even though not committed to writing until the twelfth century or later. These

44 Dumville 2008; Jorgensen 2010, 1-28. 45 The table in Mac Niocaill 1975, 40, demonstrates that the majority of the manuscripts of the Irish annals are from the 15th century or later, and none is earlier than the 11th century, even though they are assumed to have been compiled from the 8th century onwards, if not earlier (Mac Niocaill 1975, 19). 46 Shepard 2008; Hraundal 2013, 2-3. 47 Dumville 2008, 359, 361.

16 poems are often incorporated into sagas, which have their own complex relationship with the Viking Age.

Although the number of surviving texts in a Scandinavian language that can confidently be assigned to the Viking Age is relatively small, language is still an important form of continuity in the Viking diaspora, indeed language, though often ignored in these discussions, is acknowledged by some scholars to be one of the main ‘shared cultural elements’ of the Viking diaspora.48 Scandinavian settlers took their language with them to their new homes in the Viking Age, and this language

(or its later forms) continued in use in those settlements until either the settlement died out (as in Greenland in the fifteenth century) or the language and its speakers were assimilated into the majority population and its language (as happened in

England, probably around the eleventh or twelfth century).49 In Faroe and Iceland, the language has survived (and developed) continuously since the settlement period.

Even in Russia, there is evidence for Scandinavian influence on both language and naming practices.50

Various kinds of language use provide some of the best examples of continuity through the long Viking Age, an obvious example being the place-names of Faroe and Iceland that were given by the settlers and are still in use today. The maintenance of their ancestral language, in the face of other possibilities (e.g. the use of Celtic languages brought by the settlers), enabled the Icelanders to keep up close contacts with the Scandinavian homelands, and this continuing contact was an important factor in the development of Icelandic textual culture. There are a number

48 Abrams 2012, 23. 49 Townend 2002, 189, 204. 50 Svane 1989.

17 of examples of post-Viking Age contacts across the Viking diaspora that depended on a common language and textual culture, such as the literature of twelfth- and thirteenth-century Orkney, or the runic inscriptions of medieval Greenland, but the richest body of evidence comes from the extensive literature of medieval Iceland.

Sagas and the Past

Much of the prose narrative literature of high medieval Iceland has the Viking Age past as its theme. Apart from literature produced in the service of the Christian church (not discussed here) the earliest examples of this prose narrative literature are explicitly historiographical works such as Íslendingabók and Landnámabók

(discussed further in ch. 6). These and some of the kings’ sagas began to be written in the twelfth century, though many of the surviving versions are thought to have been revised or composed in their current form in the thirteenth.51 Texts in this genre are rarely stable, continuing to change and develop into the late fourteenth century or even later.52 The sagas of Icelanders (Íslendingasögur) are assumed to have been composed during the thirteenth and fourteenth centuries. While there is evidence for some of these sagas in the form of fragmentary manuscripts from the thirteenth century, many of them survive only in manuscripts of the fourteenth century or even later.53 These texts, too, continue to develop, though the variation between the versions is not usually as extensive as in the case of the historiographical works. Similarly, the sagas of ancient times or legendary sagas

(fornaldarsögur) are assumed to be a phenomenon that began in the thirteenth century, though the manuscript evidence is less clear here, and the genre was quite

51 Jakobsson 2005. 52 See e.g. Jesch 2010 on . 53 Sveinsson 1958.

18 long-lived, with a concentration of manuscripts in the fourteenth century and even later.54 It is not clear whether this fourteenth-century flourishing of different kinds of sagas is merely an accident of survival, with later manuscripts more likely to survive, or whether the literary recreation of the Viking Age past was of particular interest to authors, scribes or audiences at that time, alongside the many other literary genres concerned with other themes that also flourished then.

Once upon a time, sagas of kings, Icelanders and ancient times alike were all taken to be accurate historical accounts of the Viking Age past, and this view led to several quite serious scholarly attempts to explain how the sagas of Icelanders in particular could represent the literate culmination of a reliable oral tradition about the past.55 However the tendency of much saga scholarship in the later twentieth century and until today has been to view all three genres as, at best, literary reconstructions of that past, or using the past to mirror the present, or even outright fictions. It is acknowledged that much of the poetry in the kings’ sagas and some of the poetry in the sagas of Icelanders and ancient times may be older than the sagas in which they are preserved, having been a source for them and providing some kind of a link with earlier periods. Otherwise, much literary scholarship of the last half century or more has avoided, or downright refused to engage with, the historicity of the sagas, preferring to study them purely as literary texts.56 However, there are signs that some scholars, including archaeologists, have been interested in how they

54 Tulinius 2005. 55 E.g. Liestøl 1929. 56 E.g. Jakobsson 2013; see also Boulhosa 2005, 41.

19 might reveal aspects of their contexts of composition (in the thirteenth or fourteenth centuries) or manuscript recording (in the fourteenth century or later).57

Yet one of the most interesting things about these three saga-genres in particular, along with the historiographical works, is precisely their across-the-board obsession with the Viking Age past as their main literary theme, and this relationship between the time in which the sagas were written and that past deserves further exploration. The position taken in this book is that this pervasive connection between a high medieval literary context and the Viking Age past is an inevitable product of the long Viking Age. This long Viking Age is, in its turn, a product of the

Viking diaspora, which created lasting and wide-ranging cultural and linguistic networks and reciprocal connections between the Scandinavian homelands and the various regions settled by Scandinavians in the Viking Age. The Icelanders’ interest in their Viking Age past was thus not simply a memory of that distant past, whether accurate or constructed, but a product of those ongoing relationships long after the

Viking Age, which maintained and encouraged a sense of commonality and belonging in the present, based on a shared past.

Case Study – Vágar

An example of how these literary links across the North Atlantic and through time worked in practice can be seen in the treatment by various texts of the northern

Norwegian district of Vágar ‘Bays’ (modern Vågan), on the island of Austvågøy in

Lofoten, roughly the indented coast between the modern-day towns of Svolvær and

Henningsvær. Even today this district is the centre of a large-scale fishing industry,

57 E.g. Hjaltalín 2009.

20 and there is evidence for such activity going back to the medieval and possibly even

Viking periods.58 This particular characteristic is quite evident in the sources, but the district also had other resonances, both political and religious.

To demonstrate this requires first a detour into early Norwegian history.

Widely renowned in the sagas is a battle said to have taken place at Hjǫrungavágr

(probably modern Liavågen in Møre og Romsdal, much further south in Norway) in which Norwegian forces led by Hákon Sigurðarson, earl of Hlaðir (Lade, in

Trøndelag) defeated an invading Danish-Wendish coalition known as the

Jómsvíkingar, in about 985. Hákon’s origins and power base were in northern

Norway, north of modern , in the region known then as Hálogaland.59 The jarls of Hlaðir ruled this region quite independently despite nominal allegiance to the rulers of southern Norway, who were at times Danish. Various accounts of the battle of Hjǫrungavágr mention that one of Hákon’s supporters was a chieftain called Þórir hjǫrtr (‘Hart’) from Vágar.60 Hákon was a noted pagan, and his followers were too.61

When Hákon had been killed and Norway was ruled by the Christian missionary king

Óláfr Tryggvason, Þórir reappeared as one of the northern chieftains who attempted to resist Óláfr’s Christian mission and political ambitions, though they were ultimately unsuccessful and Þórir was killed by Óláfr.62

The religious history of the region is then obscure until over a century later when the Norwegian king Eysteinn Magnússon (d. 1122) is said to have built many

58 Perdikaris 1999. 59 Today the name Hålogaland is used in legal and ecclesiastical contexts, but politically and administratively lies in the county of Nordland. For more about Hákon, see SPSMA I, cxciii-v. 60 E.g. ÍF XXVI, 279; Hkr, p. 180. 61 Abram 2011, 127-42. 62 ÍF XXVI, 308-9, 324-5; Hkr, pp. 199, 212-13.

21 churches in different parts of Norway, including one at Vágar, to which he also granted a prebend for its maintenance.63 This seems to have marked the incorporation of the district into the medieval, Christian . The same king, Eysteinn Magnússon, along with his brothers Sigurðr and Ingi, is also said to have issued an amendment to the Frostathing law for the people of

Hálogaland, regulating their economic activities, including fishing and the fur trade.

This enactment specifically mentions that ‘every man who catches fish in Vágar’ must give five fishes to the king.64 In revising the fish tribute required from the residents of Hálogaland, the enactment implies that fishing was already a thriving industry in that region and probably increasing in importance. This importance became so great that a further royal order of 1384 names Vágar as one of the three most important trading centres of western and northern Norway, alongside and Trondheim.65

What have these rather various references to Vágar in the far north of

Norway to do with Icelanders and their sagas? Firstly, the battle of Hjǫrungavágr was a perennial favourite of the Icelandic authors of historical and pseudo-historical sagas. As well as the accounts of it in kings’ sagas such as Heimskringla, the

Icelanders produced no less than five versions of a saga known as Jómsvíkinga saga, in which an interest in the colourful exploits of the Baltic Jómsvíkingar are balanced by a West Norse perspective which is indicated by traditions recording the presence of several Icelandic poets at the battle, fighting on the side of Hákon. But

Hákon’s victory, though celebrated, was also the last of the old order, having been

63 ÍF XXVIII, 254-5; Hkr, p. 699. 64 NGL I, 257-8; ENL, pp. 404-5. 65 NGL III, 222-3; NMD, pp. 364-7.

22 achieved in part by pagan supernatural means. In the Icelandic conception of history, paganism was destined to give way to , and in their own case this happened during the reign of Óláfr Tryggvason (d. c. 1000). In Norway, his reign was short and inconclusive, but he belonged to the Icelanders as the king who had brought them to Christianity, whereas in Norway that mission was not fulfilled until the reign of his successor and namesake Óláfr Haraldsson (later St Óláfr, d.

1030).66 From an Icelandic point of view, Óláfr Tryggvason’s killing of Þórir hjǫrtr both parallels and contrasts with his more peaceful persuasion of the Icelanders to adopt Christianity.

The recalcitrance of north Norwegian pagans was not the only association the

Icelanders had with Vágar. The importance of this district as fishing station and trading centre resonates through several sagas of Icelanders, where it is presented as having had that status already around the time of the , in the ninth and tenth centuries.67 In Chapter 17 of Egils saga, the hero’s uncle, Þórólfr

Kveldúlfsson, who is based a little further south in Hálogaland, has his men fishing for in Vágar, though this is only one of his many sources of income.68

Hallfreðar saga begins in the north of Norway and mentions in Chapter 1 a boat with men on it ‘from Vágar in the north’, on which two boys escape their pursuers.69 Vágar appears in Chapters 20 and 22 of Grettis saga as a location the saga-hero Grettir visits twice while the market is on.70 The late Þorskfirðinga saga

66 On both kings, see SPSMA I, cxcviii-cciv. 67 Bertelsen 1992. 68 ÍF II, 41; CSI I, 50. 69 ÍF VIII, 136-7; CSI I, 226. 70 ÍF VII, 73, 78; CSI II, 82, 85. This market is also mentioned in an eleventh-century context in ÍF XXVII, 212, Hkr, p. 392, along with a fleet of ships known as Vágafloti.

23 also portrays (ch. 2) two visiting Icelanders sent north to Hálogaland to make money from stockfish in the time of King Haraldr hárfagri ‘Fine-Haired’.71

These two different associations of Vágar in Icelandic sources, of paganism and fishing, raise the question of its significance for the Icelandic authors and audiences of these sagas. The pagan theme and the battle of Hjǫrungavágr (fought a century after the initial settlement of Iceland) make a straightforward link to

Iceland’s heroic past. As in so much medieval Icelandic literature, the literary presentation of this past allows for the recognition of Icelandic and other heroism in a pagan context that can still be celebrated even though it was destined to be superseded by Christianity. The presentation of Vágar as a fishing station and marketplace on the other hand suggests more prosaic memories of the past. In their internal saga chronologies, two of the references to it (in Egils saga and Hallfreðar saga) are set in ninth-century Norway, before the settlement of Iceland, though involving the ancestors of the Icelandic heroes of those sagas, while Grettis saga depicts an early eleventh-century setting in which the Icelandic saga hero visits a range of places in Norway. Could these be fictionalised representations of the past based on the undoubted later status of Vágar as a fishing and trading centre in the thirteenth and fourteenth centuries when these sagas were written?72 Or do they represent the medieval Icelanders’ memories of their pre-settlement history in

Norway? The archaeological evidence suggests that the answer is probably both, the two reinforcing each other.

71 ÍF XIII, 182; Þorsk, p. 134. 72 Perdikaris 1999, 395-6.

24

It has been thought that Iceland’s own trade in exporting stockfish did not take off in a big way until the late thirteenth or even fourteenth century, though more recent zooarchaeological analyses suggest rather that it could have been underway by 1200.73 The creation of wealth from the Lofoten fishing grounds, on the other hand, was not new in the thirteenth century, though how far back the lucrative large-scale export of fish actually went is more difficult to establish. In

1974, Thorleif Sjøvold could declare, based on the paucity of fishing gear found in the graves and stray finds of Arctic Norway, that ‘fishing was of far less importance to the Late Iron Age population that would have been expected’, but this impression now seems ill-founded.74 The Norwegian king’s interest in Vágar in the early twelfth century, as mentioned above, suggests that trade, quite probably in fish, was already important by then. Even earlier, the powerful Viking Age chieftain who had his seat at Borg, on the neighbouring island of Vestvågøy, is thought to have acquired his immense wealth through the exchange of local surplus or hunting products paid to him as tribute.75 The archaeological evidence of boat-houses in the area, and more recently from the stable isotope analysis of fish-bones found in

Hedeby, suggests that some of this local surplus was fish, and thus that fishing for long-distance export, particularly of stockfish, could have taken place in this region as early as the ninth century.76 A recent isotope analysis of samples taken from 33 burials in the north of Norway shows an increase in the consumption of marine protein over time, and also that certain individuals changed their dietary habits later

73 Karlsson 2000, 106-10; Amundsen et al. 2005; Keller 2010, 2. 74 Sjøvold 1974, 302-3, 352. 75 Näsman and Roesdahl 2003, 290. 76 Johansen 2003, 28-9; Barrett et al. 2008, 857-8. Perdikaris 1999, 398-9 argues that the North Norwegian stockfish trade in the Viking Age was for the purposes of ‘chieftainly prestige’ rather than for ‘monetary profit’ as later on. See also Bertelsen 1992.

25 in life, suggesting a move to a coastal area.77 Both phenomena are observed in the

Viking Age (before 1030) and are consistent with households whose economy was based partly or primarily on the exploitation of marine resources, in connection with the expanding stockfish industry.

Landnámabók lists a number of Icelanders who came from Hálogaland, or even Lofoten, and it is likely that their descendants retained a memory of their home district and its source of wealth.78 Several of the anecdotes told of these immigrants from northern Norway relate to their generosity and ability to provide food, such as

Geirríðr who sat outside her house inviting passers-by in to eat, or Þuríðr sundafyllir who was called ‘Sound-Filler’ because of her magical ability to conjure fish into the waters of Hálogaland at times of famine.79 Þengill mjǫksiglandi (‘Frequently-Sailing’) may have been named thus because of his fishing exploits.80 These anecdotes seem to preserve truthful memories of the fish-basket that was Hálogaland. Any such memories of the settlement period would have been strengthened by continuing contacts with the Norwegian homeland at the time the sagas were being written and in the context of the continuing flourishing of the fishing industry there. The strong paganism of Hálogaland is also reflected in these anecdotes. As well as Þuríðr with her magical ability to conjure up fish, a certain Eyvindr is said to have made sacrifices, and Óláfr tvennumbrúni ‘Double-Eyebrowed’ is said to have been hamrammr mjǫk ‘a great shape-changer’ and to have been buried, pagan fashion, in a mound.81 Unlike the memories of the fishing industry, such anecdotes are less

77 Naumann et al. 2014. 78 ÍF I, 127, 186, 244, 257, 272-3, 282, 332, 348-9, 376; BoS, 46, 69-70, 95, 99, 103-4, 107, 126, 130-31, 140. 79 ÍF I, 127, 186; BoS, pp. 46, 69-70. 80 ÍF I, 272; BoS, p. 103. 81 ÍF I, 273; BoS, p. 140; NIP, col. 389.

26 likely to have been strengthened by continuing contacts with the Norwegian homeland, which was by then no longer pagan. Rather, they plug into the literary fascination with the north of Norway that is evident in the fornaldarsögur.

By way of contrast to the Icelandic perspective, we have a glimpse of Vágar from that of a thirteenth-century Norwegian, the anonymous author of Konungs skuggsjá, an instructional text for an ambitious young man who aspires to be a merchant. For this author, Vágar, and its northern neighbour Andarnes (modern

Andenes), are characterised by neither paganism nor fishing but by their latitude. To this up-to-date and scientifically-minded observer, instructing the future seafarer,

Vágar is simply (and slightly inaccurately) the land of midday stars in winter and the midnight sun in summer.82 The Icelanders’ view of the same place, focused on its past paganism and its ongoing status as a major fishing port, is on the other hand an intermeshing of past and present, homeland and new land, which is characteristic of the literature of the Viking diaspora.

Understanding the Viking Diaspora

Defining Diaspora

The discussion so far has suggested some of the ways in which natural, artefactual and linguistic evidence demonstrate strong links between the Viking Age and what followed it. Several of the examples have also shown how these links depend on continuing cultural contacts between different parts of the Viking world, not always

82 KS, pp. 12, 131-2; KM, pp. 98-9.

27 including the homeland. The question to be explored now is whether these connections across time and space justify the concept of a Viking diaspora, emanating from Scandinavia, but extending both east and west, and developing into a series of complex, reciprocal networks between the homelands and the settlements that lasted for differing periods of time. In particular, an approach based on texts provides a diasporic interpretation that can then be tested against other types of evidence. This approach is outlined below and then exemplified in the following chapters.

Before going on to discuss the Viking diaspora, it is worth pausing to remember that the word ‘diaspora’ was originally applied to the dispersal of the Jews from their homeland in Palestine and it is sometimes argued that it is inappropriate to use it for other historical situations.83 However, recent work by Shlomo Sand, although he does not use the word ‘diaspora’ very much, challenges the entrenched idea that the spread of Judaism was due to a traumatic dispersal of peoples, involving massive migration, and concludes rather that it came about through a dynamic process of conversion leading to the spread of the religion outside of

Palestine over a period of time.84 However, whether or not the spread of Jews throughout the world is recognised as the prototypical diasporic experience, the term

‘diaspora’ is now used in a wide variety of contexts, both historical and contemporary, although usually still in contexts involving the movements of people.85

83 OED, ‘diaspora’; Safran 2005; Abrams 2012, 19 n.6. 84 Sand 2010, esp. 188-9. 85 E.g. Kalra et al. 2005, 8-10; Abrams 2012, 19.

28

Steven Vertovec’s distinction between ‘diaspora’ and ‘migration’ has already been cited in Chapter 1 but is repeated here for convenience. He sees

... migration as physical movement, resettlement and re-establishment of key social institutions; diaspora as the consciousness of being connected to the people and traditions of a homeland and to migrants of the same origin in their countries; and transnationalism as the practices of exchange of resources, including people, across the borders of nation states. This implies that migration can occur without diaspora and transnationalism, but the two last-mentioned activities are always a result of migration.86

The third stage of transnationalism is less relevant in the present context, as it presupposes the existence of well-defined and recognised nations, although these did emerge during the period under consideration here, and some of the later diasporic connections identified below might just as well be seen as transnationalism.

The reciprocal relationship with the homeland which characterises diaspora in contrast to migration is further emphasised by Kalra et al. as follows:

... diaspora more often than not evokes two social spheres of interaction – the place of residence and the place from which migration has occurred. ... It is the ongoing political, economic, social and cultural ties between multiple institutionalized spaces that characterize diaspora.87

This reciprocal aspect is often more than just binary (as recognised by Vertovec) and this complexity, which seems particularly relevant to the Viking Age and its aftermath, was formulated by the anthropologist James Clifford as the ‘[d]ecentered,

86 Mirdal and Ryynänen-Karjalainen 2004, 8. 87 Kalra et al. 2005, 3.

29 lateral connections’ of diaspora, which ‘may be as important as those formed around a teleology of origin/return.’88 Thus a full understanding of a diaspora has to move away from a focus just on the homelands and colonies, and recognise the importance of reciprocal, and ongoing, links between the different regions of the diaspora.

Diaspora theory was introduced to Viking Studies with the creation of the

AHRC-funded Viking Identities Network in 2006.89 Since then the term has been widely adopted in the field but with little justification or discussion. Steinunn

Kristjánsdóttir, for example, gives a brief definition of diaspora (‘a population sharing common ethnic identity, but that left freely or forced their settled territory and became residents in new areas’) but without further exploration of the implications of this, nor any justification of her use of the term to discuss religion in early

Iceland.90

There has been just one detailed discussion of the validity of the concept of diaspora for Viking Age studies, offered recently by Lesley Abrams.91 Abrams cites

Robin Cohen’s proposed attributes of diaspora (see more on this below), but does not address them systematically or in detail, only noting that ‘Some apply quite readily to the Viking Age, others are more problematic’.92 Instead she proposes her own ‘model of Viking-Age society that takes in the Scandinavian homelands and overseas settlements’, which conceives the Viking diaspora to a great extent as a

88 Clifford 1994, 306; see also Kenny 2013, 39. 89 Jesch 2008b explains the rationale of the network and appears to have been the first published work to suggest the usefulness of the term ‘diaspora’ to the Viking Age and its aftermath; see also Abrams 2012, 17. 90 Kristjánsdóttir 2011, 431. 91 Abrams 2012. 92 Abrams 2012, 19-20.

30 series of royal or elite courts in various kinds of contact with each other.93 Abrams concludes that ‘for a period the dispersed Scandinavian communities of the Viking

Age acted like a diaspora, retaining, synthesizing, and expressing a sense of collective identity and constructing a common cultural discourse, while new circumstances generated innovations and developments which flowed back and forth between them’.94 While this is not incompatible with the general argument presented here, it is a narrower vision, probably too cautious (‘for a period’, ‘acted like a diaspora’), and the model is heavily dependent on evidence from the British Isles and continental Europe (not really considering the ‘overseas settlements’ where there were no ‘host communities’) and especially on material culture, as well as being restricted chronologically to the Viking Age proper and largely to elites. It is argued here rather that the concept of ‘the Viking diaspora’ is most useful when applied in the context of the full range of evidence and the extended chronology of the long and broad Viking Age as outlined above.

The most thorough, and seminal, account of diasporas is by the sociologist

Robin Cohen. He proposed that, normally, ‘diasporas exhibit several or most of the following features:

1. dispersal from an original homeland, often traumatically; 2. alternatively, the expansion from a homeland in search of work, in pursuit of trade or to further colonial ambitions; 3. a collective memory and myth about the homeland; 4. an idealization of the supposed ancestral home; 5. a return movement or at least a continuing conversation;

93 Abrams 2012, 21. 94 Abrams 2012, 38.

31

6. a strong consciousness sustained over a long time; 7. a troubled relationship with host societies; 8. a sense of co-responsibility with co-ethnic members in other countries; and 9. the possibility of a distinctive creative, enriching life in tolerant host countries.’95

Cohen’s analysis is based on recent history but many, in fact most, of these features can be demonstrated in the movements of the Viking Age and the societies and cultures that developed from these movements, sufficient to qualify them as a diaspora.

As noted above, Abrams did not address this paradigm point by point, and for her ‘[t]he most obvious practical difficulty in applying Cohen’s diaspora paradigm ... is inequality of information. It is very difficult to conceptualize society in the Viking

Age, because our evidence is both slight and complicated.’96 As a historian, Abrams is very concerned with the problem that our ‘historical sources’ are written by outsiders, or later than the Viking Age, and that material culture has its own problems of interpretation. Her model of the Viking Age diaspora is quite firmly based in contemporary evidence from the Viking Age itself, rather than in a more generous acceptance of evidence from a longer period and a broader geographical context. The latter is the method proposed here, which involves accepting the limitations of this evidence, but also recognising that it may still contain significant and revealing patterns, the ‘remarkable patterns of continuity’ mentioned at the beginning of this chapter. With such a broader perspective in both time and space, it

95 Cohen 2008, 161-2; see also Abrams 2012, 19; and a similar list in Safran 2005, 37. 96 Abrams 2012, 20.

32 is possible to address Cohen’s points one by one in relation to a putative Viking diaspora, as seen in a range of selected examples from different places, periods, and based on different types of evidence. These examples are meant to be illustrative rather than exhaustive, and some themes arising from them will be considered in more detail in the following chapters.

‘Dispersal from an original homeland, often traumatically’

This feature is instantly recognisable in the persistent Icelandic myth that its first settlers left Norway to escape the tyranny of King Haraldr hárfagri ‘Fine-Haired’.

Such a motif is widely found in those Sagas of Icelanders that cover the settlement period, as expressed in the words of the Norwegian chieftain Ketill flatnefr ‘Flat-

Nose’ in Laxdœla saga (ch. 2):97

‘Sannspurðan hefi ek fjándskap Haralds konungs til vár; sýnisk mér svá, at vér munim eigi þaðan trausts bíða; lízk mér svá, sem oss sé tveir kostir gǫrvir, at flýja land eða vera drepnir hverr í sínu rúmi.’

‘I have heard true reports of King Haraldr’s enmity towards us; it seems to me that we will not experience safety from that direction; it appears to me that there are two choices open to us, to flee the country or to be killed, each in his turn.’

Ketill’s sons want to go to Iceland, though the old man is less keen, preferring

Scotland. It is characteristic that those who were dispersed from Norway in this way often ended up in a variety of places. Thus in Eyrbyggja saga (ch. 1) the noble men

97 ÍF IV, 4; CSI V, 2. See also Egils saga ch. 25 (ÍF II, 65; CSI I, 61), Gísla saga ch. 1 (AM 445c version, ÍF VI, 3), Grettis saga ch. 3 (ÍF VII, 6; CSI II, 51). In Kormáks saga ch. 2 (ÍF VIII, 204-5; CSI I, 180), the tyrannous king is Haraldr’s son Eiríkr blóðøx ‘Blood-Axe’ and in Þorsteins saga hvíta ch. 1 (ÍF XI, 3; CSI IV, 303), Þorsteinn goes to Iceland after the death of his father who had to move within Norway because of the tyranny of Earl Hákon. See also Boulhosa 2005, 172-3, 176, 228-32.

33 who left their ancestral estates are said variously to have gone across the Keel to

Sweden, or west across the sea to the Hebrides or Orkney, before going on to

Iceland (ch. 6).98 Similarly, Færeyinga saga gives Haraldr’s tyranny as the explanation for the settlement of the ‘and other uninhabited lands’.99

Even those chieftains who were friendly with Haraldr, such as Ingimundr in

Vatnsdœla saga (chs 8-12), are presented as having moved to Iceland at the time of and in the context of the unrest in Norway.100 The traumatic dispersal westwards from Norway is also reflected in Landnámabók, which regularly notes that certain settlers left for political reasons.101

Similar traumatic dispersals can be found in other literary traditions of the

Viking diaspora. The Middle English romance Havelok the Dane (from around 1300), written for a Lincolnshire audience still aware of its Danish heritage, tells the story of

Havelok, the son of the Danish king Birkabeyn, who as a small child is imprisoned after his father’s death along with his sisters.102 Their supposed guardian kills the sisters, but Havelok escapes through the help of a kindly peasant, Grim, who eventually takes his whole family, including Havelok, to England, where he founds the settlement known as Grimsby. Havelok eventually marries an English princess and becomes king of England. Written from a later medieval perspective, the poem’s primary purpose is to celebrate the harmoniously dual Anglo-Saxon and Viking

98 ÍF IV, 3-4, 11; CSI V, 131. 99 IF XXV, 3; ThoG, p. 19. 100 ÍF VIII, 22-36; CSI IV, 11-17. As well as Ingimundr, the sons of Haraldr’s closest follower Rǫgnvaldr Mœrajarl end up in either Iceland or Orkney. See also Boulhosa 2005, 179-82, on positive attitudes to Norway in Finnboga saga. 101 ÍF I and BoS: Sturlubók version chs 29, 84, 85, 112, 134, 135, 139, 144, 149, 154, 156, 159, 161, 166, 225, 241, 267, 341, 344, 356, 359, 371, 378, 392 and similarly in other versions. Some, though not all, of these examples reflect the influence of the sagas on this text. As in the Sagas of Icelanders, there are some settlers (e.g. chs 284, 309-10) who went to Iceland in a state of friendship with the Norwegian king. 102 Hav.

34 heritage of the English nation, and to acknowledge the full assimilation of the

Danish-origin inhabitants of Lincolnshire into this nation.103 Havelok’s trajectory from prince to pauper and back again is a common romance motif, but while it should not be taken too literally, the story does plug into local memories of the Danish migration to Lincolnshire. As an explanation for this migration, the tyranny of the

Danish ruler presented in Havelok parallels the role of Haraldr hárfagri in Norway, and suggests that the story of traumatic emigration was useful elsewhere in the

Viking diaspora than Iceland.

‘Alternatively, the expansion from a homeland in search of work, in pursuit of trade or to further colonial ambitions’

Those settlers discussed above who are said to have gone to Iceland with the blessing of King Haraldr hárfagri might be said to have gone with ‘colonial ambitions’. Similarly, Orkneyinga saga (ch. 4) has King Haraldr give Orkney and

Shetland to his lieutenant Rǫgnvaldr, Earl of Møre, in compensation for the death of

Rǫgnvaldr’s son while serving Haraldr.104 The first few chapters of the saga show some of the difficulties Haraldr had in establishing control over this colony. A colonial context is also implied in the account of the Russian Primary Chronicle telling how the ‘Varangians from beyond the sea’ first imposed tribute on a wide range of inhabitants in the east in 859, were at first resisted, and then were called back to reign over the unruly region.105 England’s ‘Second Viking Age’ culminated in the crowning of the Danish king Knútr as sole King of England in 1017. While there were many factors which brought this about, the Danelaw, a large region where many of

103 Turville-Petre 2001, 348-54. 104 ÍF XXXIV, 8; OS, pp. 138-9. 105 RPC, p. 59; see Franklin and Shepard 1996, 38-9.

35 the population were of Scandinavian origin, provided a useful launching-pad for the military activities that ensured Knútr’s final success, making him into ‘the most successful of all pre-Conquest rulers in Britain’, but also one whose ‘dominion and influence extended over much of the northern world’.106

Nevertheless, it is doubtful whether a culture that did not have a developed concept of the state could be said to have colonial ambitions. In any case, colonial ambitions could only be realised in particular circumstances and are perhaps rather a by-product than a cause of the Viking diaspora. In general, land-hunger or the opportunity to trade are more commonly adduced as important factors in the

Scandinavian expansions to various parts of the world.107 While trade need not lead to permanent settlements, in which case the applicability of the term ‘diaspora’ is questionable, land-hunger could and would bring about diasporic settlements.108 The

Anglo-Saxon Chronicle famously states that in 876, after a period of Viking raids in many parts of England, their leader ‘Halfdan shared out the land of the

Northumbrians, and they proceeded to plough and to support themselves’ and similarly in 877 that they shared out some of Mercia.109 Certainly the whole of

Landnámabók conceptualises the settlement of Iceland as the creation of a farming society from scratch, with opportunities for all, and within six decades.110 This variety of explanations for the Viking diaspora, both ancient and modern, undoubtedly reflects a multiplicity of causes for it.

106 ASC, p. 97; Lawson 2004, 82, 196. 107 See e.g. Barrett 2010, 291-5, though he argues against a strict demographic or economic determinism as causes of the Viking Age; also Dugmore et al. 2007b, 15; Abrams 2012, 28. 108 Both Abrams (2012, 28) and Cohen (2008, 83-104) use the term ‘trade diaspora’, though this involves stretching the concept in a new direction which is not necessarily helpful. 109 ASC, p. 48. 110 ÍF I, esp. 396; BoS, p. 146.

36

‘A collective memory and myth about the homeland’

The Norwegian origin of the Icelanders is stated very prominently in Chapter 1 of

Íslendingabók, which also makes clear (ch. 2) that Iceland’s first law code was brought from Norway, soon after the settlement.111 This firmly establishes the

Icelandic polity as an offshoot of the Norwegian one. The Russian Primary Chronicle notes that ‘[t]he present inhabitants of Novgorod are descended from the Varangian

[= Scandinavian] race, but aforetime they were Slavs.’112 Both of these represent different collective memories about homeland origins. The collective nature of the

Icelanders’ memories about their homeland origins is illustrated on a larger scale in

Landnámabók, which regularly specifies a precise place of origin for the first settlers, usually though not always in Norway, as well as in those sagas which begin with their protagonists’ origins in Norway.113

In contrast, Orkneyinga saga (chs 1-4) presents a historicised myth about the

Norwegian origins of the Orkney earldom, rather than a collective memory of individuals and their places of origin. In an elaborate story, Norway is envisaged as having been conquered in mythological times by two brothers, Nórr who rules the mainland and Górr who is king of the isles and skerries. The earls of Orkney are ultimately descended from Górr.114 This myth establishes a link between the two countries but also emphasises their separateness. Norway is figured not so much as an ancestral home but more as a country of the same origin as Orkney which experienced parallel political developments. Preben Meulengracht Sørensen has

111 ÍF I, 4-7; BoI, p. 4. 112 RPC, p. 60. 113 This will be discussed further in ch. 6. 114 ÍF XXXIV, 3-8; OS, pp. 135-9.

37 argued that this Orcadian origin myth expresses ‘a specific Nordic self-esteem and pride’, indeed that it shows that the earls of Orkney are ‘more Norwegian than the kings of Norway’.115 However, the provenance of this myth is not clear, whether it is an Icelandic construction or an Orcadian interpretation of the past, nor is it clear whether it was part of the saga from its inception. Indeed the myth may have had more to do with political conditions in both places in the thirteenth or even fourteenth centuries than with any collective memories of the homeland. Even so it would show that origin-stories were still a matter of interest and possibly debate at that time.116

‘An idealization of the supposed ancestral home’

It is debatable whether the Icelanders actually idealised their ancestral home in

Norway, but they certainly took a great interest in it and its history. Theodoricus, a

Norwegian monk who wrote a Latin in the late twelfth century, explicitly notes that he has much of his information from the well-informed

Icelanders.117 Similar sentiments were expressed by the Danish historian Saxo

Grammaticus, writing in Latin around 1200, and of course the Icelandic kings’ sagas also included sagas of the kings of Denmark.118 This suggests that the Icelanders’ concept of their ancestral home, though focused on Norway, encompassed the larger cultural zone that we call Scandinavia. The heroic prehistory of Scandinavia is celebrated in many of the fornaldarsögur, which are mainly set in the northlands in a

115 Meulengracht Sørensen 1993, 219, 221. 116 Rowe 2005, 316-36, discusses these chapters and related texts in detail. 117 TM, pp. 1, 5, 55. 118 Saxo, p. 5.

38 time before the settlement of Iceland, but often involving characters who are figured as the ancestors of certain Icelanders.119

Another aspect of this idealisation of the ancestral home can be glimpsed in naming habits. Place-names in the new settlements which seem to have been transferred from the homeland, rather than given afresh, suggest that the settlers had sufficiently good memories of their old homes to want to perpetuate their names in the new. Icelandic place-names such Þelamörk, Finnmörk, Katanes and

Sigtún are most likely to have been named after Telemark and in Norway,

Caithness in Scotland and Sigtuna in Sweden respectively, rather than being coined because of any local considerations.120 Since few of these names are recorded in very early documents, it is however possible that they were coined some time after the initial settlement, on the basis of continuing contact with or knowledge of these places.

‘A return movement or at least a continuing conversation’

The reciprocal traffic between the diasporic lands and Scandinavia is well- documented and certainly qualifies as a ‘continuing conversation’. Landnámabók, in noting that Iceland was discovered and settled from Norway, gives a description of how to sail to Iceland, with a starting point at the prominent landmark of Stadlandet in , locating the two countries in an ongoing geographical relationship.121 The description is expanded on the basis of personal knowledge in the version of Landnámabók written by Haukr Erlendsson, who spent much of his life

119 Quinn et al. 2006, esp. 278, 294. 120 Lárusson 1939, 64. For a more cautious view, see Sigmundsson 2009, 227-8. 121 ÍF I, 32-3; BoS, p. 16.

39 in Norway in the early fourteenth century. The Sagas of Icelanders show many a young man going to Norway for three or more years in order to establish himself, as well as some merchants who are active there even longer.122 The kings’ sagas

(especially, but not only, Morkinskinna) regularly retail anecdotes about Icelanders

(especially poets) at the courts of the eleventh-century kings of Norway.123 Similarly,

Norwegians are regularly to be found in Iceland, in a variety of sagas, and occasionally even represented as living there.124 These Norwegians are presented as strangers, recognisably not Icelanders, and yet they are not as strange as any other non-Icelandic group would be, and there are a lot of them in the texts.

The place of Norway in the Icelandic consciousness can be deduced from the laws preserving various rights, particularly inheritance rights, of Icelanders in

Norway. These legal rights come with reciprocal responsibilities to the king of

Norway from Icelanders, and particular rights for the Norwegian king in Iceland.125

The reciprocity of the provisions is clear, and is equally clearly derived from the

Norwegian ancestry of the Icelanders, a fact used in the Icelanders’ negotiations for these rights. The ‘continuing conversation’ was taken to its logical conclusion when the Icelanders submitted to the king of Norway in 1262-4. Patricia Boulhosa has argued that this was ‘the result of a continuous development’ in which Icelanders negotiated their way towards that relationship, a relationship which encompassed

122 E.g. Gunnlaugs saga ch. 5 (ÍF III, 64), Bjarnar saga hítdœlakappa ch. 2 (ÍF III, 113), Laxdœla saga ch. 40 (ÍF V, 114-15), Hallfreðar saga ch. 4 (ÍF VIII, 150), Kormáks saga ch. 18 (ÍF VIII, 260), Njáls saga ch. 2 (ÍF XII, 11). 123 Msk, pp. 171-9, 187-94, 206-8, 211-15, 219-20, 222-3, 243-52, 255-61. 124 Callow 2004, esp. 330-31; Jakobsson 2007. 125 Boulhosa 2005, 85-6.

40 both positive attitudes towards the Norwegian king and the maintenance of certain aspects of the Icelandic way of life.126

The relationship with Norway was just as important in Orkney. The history of the earls of Orkney suggests that their power could only be maintained with the support of Norway, as noted in Chapter 1, above, in the case of Þorfinnr Sigurðarson in the eleventh century. Orkney’s most notable twelfth-century earl, Rǫgnvaldr Kali

Kolsson (d. 1158/9) was the offspring of an Orcadian mother and a Norwegian father, and grew up in Norway, and his near-contemporary and fellow-poet Bishop

Bjarni Kolbeinsson had a similar background.127 Marriage alliances with powerful families in Norway took place in Iceland just as in Orkney.128

In England, events leading up to its conquest by the Danish king Knútr in

1016 have been called England’s ‘Second Viking Age’. In fact, the archaeological evidence suggests three rather different stages to the contacts between England and

Denmark in particular.129 There was first an early stage of raids and attacks, followed by settlement in England. In neither of these stages is there much evidence of these interactions back in Denmark. However, the third stage, which starts under the reign of Sveinn tjúguskegg ‘Forkbeard’ in the late tenth century, brings in a whole new era of traffic and exchange back and forth across the , particularly in objects which draw on both traditions. Although doubtless facilitated by Knútr’s joint rule of England and Denmark, this process continued well beyond his

126 Boulhosa 2005, 209-13. 127 SPSMA II, 575, and I, 954. 128 Jakobsson 2007, 148. 129 Pedersen 2004.

41 death in 1035, maintained by ‘family ties and relations, the potential for traide, as well as the ecclesiastical organisation’.130

‘A strong ethnic group consciousness sustained over a long time’

Much archaeological ink has been spilt in the last few decades on the question of whether and how material culture expresses or constructs ethnic or other group identities.131 The nature of the association between material culture and any form of

‘ethnic consciousness’ is a complex matter not easily reduced to a simple equation.

Easier access to ‘consciousness’ is provided by language and how it is used.

The group consciousness of the Viking diaspora is perhaps best expressed in a term used of its common language, the dǫnsk tunga ‘Danish tongue’. This phrase encapsulates the linguistic unity of Scandinavia, perhaps because Danish territory would be where any Scandinavian returning home from the south would first feel linguistically at home. The phrase is recorded in the early eleventh century in a poem by the Icelander, Sigvatr Þórðarson, celebrating the return from the European continent of Óláfr Haraldsson to take up the kingship of Norway.132 It is also used in one version of an early saga about King Óláfr Tryggvason, probably composed just before 1200, though describing that king’s Christian mission two centuries earlier.133

The context is the difficulty of this mission since the foreign missionaries were ignorant of or inexperienced in using the dǫnsk tunga. Since the text credits Óláfr with converting Shetland, Orkney, Faroe, Iceland and Greenland, it has to be

130 Pedersen 2004, 67. 131 The classic study is Jones 1997; for the Viking Age, see Stig Sørensen 2009 and Gräslund 2009, 137-9, both with further references. 132 SPSMA I, 555-6. 133 ÍF XXV, 271; SOTOS, p. 102.

42 assumed that the phrase refers to the forms of language spoken in all of those places, and shows an awareness of their close kinship with the languages of the

Scandinavian homeland. The common Scandinavian nature of dǫnsk tunga is clearly expressed by Snorri Sturluson, thirteenth-century Icelandic poet, historian and mythographer. In his Edda, a work of mythology and poetry, he twice uses the phrase, once specifically equating it with all the Northlands where a common language is spoken (as he also does in Heimskringla), and once contrasting it with the languages spoken in Saxland (roughly the northern part of modern ) and England.134 In England itself, it seems that the word denisc ‘Danish’ was used of all Scandinavian peoples, before usage gradually became more specific.135

In Chapter 17 of Knýtlinga saga, King Knútr is said to have been ‘the most powerful king and the one with the most land á danska tungu’.136 While this might not be surprising in a king who was from Denmark itself, this comment comes immediately after an account of all the places in which he ruled, namely Denmark,

England, Norway and even (probably incorrectly) Scotland. This is then followed by an account of Knútr’s pilgrimage to Rome, and his establishing of hostels along the way for all of those who went there who were af danskri tungu. Even more generous was the Danish king Eiríkr Sveinsson (d. 1103) who, on his pilgrimage to Rome, left money for all pilgrims who spoke danska tungu to be given free wine and lodging at the hostel in Lucca (ch. 74).137

134 SnESkskm, pp. 52, 80; SnE, pp. 107, 129; ÍF XXVI, 3; Hkr, p. 3. 135 Pons-Sanz 2013, 221-2. 136 ÍF XXXV, 123; KnS, p. 40. 137 ÍF XXXV, p. 220; KnS, p. 110.

43

The most revealing use of dǫnsk tunga is however in the thirteenth-century

Icelandic laws, where knowledge of this language is a prerequisite of full participation in the legal system. Thus, someone who has not learned to speak it in childhood cannot be nominated as a member of the court until he has spent at least three years in Iceland.138 Special inheritance rights are given to heirs who speak ‘our language’, whereas those whose languages are less familiar, such as Englishmen, have to fulfil more stringent criteria.139 There are several other provisions in the laws which show a clear sense of a common tongue with ‘foreigners’ who are Danish,

Swedish or Norwegian.140 This sense of a common tongue appears to have lasted in

Iceland at least until the end of the fourteenth century.141

There are other words expressing a Scandinavian group consciousness. An adjective frequently used in the Sagas of Icelanders is norrœnn, which normally means ‘Norwegian’, as opposed to ‘Icelandic’, as in Chapter 6 of Valla-Ljóts saga, in which a group of men newly arrived in the north of Iceland on a ship are identified as suma ... íslenzka, en suma norrœna ‘some ... Icelandic and some Norwegian’.142

In this sense, the adjective is equivalent to the noun austmaðr ‘person from the east’, which in an Icelandic context usually means a Norwegian.143 However, these meanings are contextual and, in episodes set outside Iceland, norrœnn can just mean ‘Scandinavian’ (or more probably ‘West Scandinavian’). Thus, in Chapter 53 of

Egils saga, the unit of the English king Aðalsteinn’s army at the battle of Vínheiðr is described as including allir norrœnir menn ‘all the Norse men’ who were at the

138 GrágásK I, 38; LEI I, 53; see also LEI II, 388. 139 GrágásK I, 229; GrágásSt, pp. 74-5, 95-6; LEI II, 11, 244-5, 247. 140 GrágásK I, 172, 244, II, 198; GrágásSt, pp. 74-5, 95-6; LEI I, 160; II, 24, 214, 244-5. 141 Jakobsson 2007, 151; see Flat I, 296 and SPSMA II, 454-5; VII, 27-8, 566. 142 ÍF IX, 250, CSI IV, 140. 143 E.g. ch. 1 of Ljósvetninga saga, ÍF X, 5; CSI IV, 194; ONP, ‘austmaðr’.

44 battle, and they are equipped with ‘Norse’ shields and armour.144 This unit is led by

Egill’s brother Þórólfr, both of them of course Icelanders. Similarly, the noun norrœna ‘Norse’ is occasionally used of the common language, rather than of the specifically . This, too, is used in contexts where the commonality of Scandinavian contrasts with other languages. The German character

Tyrkir switches from speaking á þýzku ‘in German’ to speaking á norrœnu ‘in Norse’ in Chapter 4 of Grœnlendinga saga, and in Chapter 64 of Eyrbyggja saga an

Icelandic merchant’s ship is driven to a curious place southwest of Ireland where everyone seems to be speaking Irish, except for their leader who addresses the arrivals á norrœnu ‘in Norse’ (and who is thought to have been an Icelander, though the saga narrator is cautious on this point).145

There is plenty of evidence for a recognition of different nationalities within the larger ethnic group consciousness. The adjective norrœnn was also borrowed into English, where it is used to contrast ‘Norwegian’ with ‘Danish’, at least in northern dialects, where they may have been more aware of these differences.146 A well-known episode in Eyrbyggja saga (ch. 50), shows the arrival in Iceland of a merchant ship from , on which are both Irish and Hebridean people, en fáir norrœnir ‘but few Norwegians’.147 Here, the Icelandic context suggests the

‘Norwegian’ meaning for norrœnn, but it is also of interest to note that one of the

Hebrideans, Þórgunna, has a Norse name, suggesting that in this episode the

Hebrideans and the Irish may actually have been Norse-speaking Scandinavians based in those places. Similar distinctions between ethnicity and place of residence

144 ÍF II, 136-7; CSI I, 96 translates this word as ‘viking’. 145 ÍF IV, 178, 180, 252; CSI I, 22 (ch. 3), V, 216. Compare ch. 21 of Laxdœla saga (ÍF V, 54; CSI V, 27). 146 Pons-Sanz 2013, 221-2. 147 ÍF IV, 137; CSI V, 195.

45 may lie behind certain place-names in England, for example several places with names of the type Irby/Ireby in the north-west, which are usually assumed to be named after a Hiberno-Norse settler rather than an actual Irishman.148

Linguistic distinctions are not always the same as ethnic distinctions either.

The mid-twelfth-century author of the First Grammatical Treatise identifies himself as belonging to the Íslendingar ‘Icelanders’; while most commonly referring to their and his language as várt mál ‘our language’, he also once calls it dǫnsk tunga.149

Snorri Sturluson, in his history of the kings of Norway, makes an interesting observation on the place-names of Northumbria, in England, noting that they are given á norrœna tungu ‘in the Norse tongue’, having just noted that the region was densely populated by Norðmenn (possibly ‘Scandinavians’ in this instance) and that

Danir ok Norðmenn ‘ and (?)Norwegians’ had harried and ruled there.150 The multiplicity of terms used here suggests a more general awareness of difference within the overarching linguistic unity of the Scandinavians.

‘A troubled relationship with host societies’

The depredations of Viking raiders in Britain and Ireland, and on the European continent, are well-documented from contemporary sources and have been covered extensively in previous scholarship.151 That the violence was not all one way is demonstrated by two important recent archaeological discoveries from the south of

England, the mass graves from Ridgeway Hill, Dorset, and St John’s College, Oxford.

These represent two separate massacres of young Scandinavian men (interpreted as

148 Fellows-Jensen 1985, 16-17. 149 FGT, pp. 208-9, 212-13. 150 ÍF XXVI, 153; Hkr, p. 98. 151 E.g. Keynes 1997; Nelson 1997; Holman 2007; Dumville 2008; Ó Corráin 2008;

46 raiding parties) in the tenth or early eleventh centuries.152 In both cases the

Scandinavian origins of the dead were established through stable isotope analysis.

While the historical context for the Ridgeway massacre is uncertain, it was tempting to see the Oxford bodies in the context of a massacre of Scandinavians which, according to the Anglo-Saxon Chronicle, took place in Oxford on St Brice’s Day, 13

November 1002. 153 In this case we know that the perpetrators were acting on the orders of the English king, and that the victims were said to have been denisc

‘Danish’, but their age and gender, or even number, is less clear. However, the dating evidence, though not conclusive, tends to rule out the Oxford burials as being the result of the St Brice’s Day massacre, so that they represent a different event.

These three events together provide extraordinary evidence for the extent of violence towards Scandinavians in the south of England within a period of fifty years at most.

It is not clear when the inevitable violence of raiding armies (or retaliations against them) turns into a ‘troubled relationship’, nor how much violence was needed for Scandinavian settlers to establish themselves in regions that were already inhabited. There has for instance been a long-standing discussion about the fate of the indigenous (Pictish) inhabitants of the Northern Isles, with some scholars arguing that they experienced what could only be described as genocide, while others maintain that the evidence rather suggests processes of transition and acculturation than conflict.154 Even in Iceland, where there was no established indigenous population, there were memories of conflict with some Irish clerics who

152 Pollard et al. 2012; Loe et al. 2014. 153 ASC, p. 86. 154 See for instance the contrasting interpretations of Smith 2001 and Bäcklund 2001.

47 were in situ when the first Scandinavians arrived, according to Íslendingabók (ch.

1).155 These clerics, however, simply left rather than resisting the incomers because, as Christians, they did not wish to co-exist with heathens.

‘A sense of co-responsibility with co-ethnic members in other countries’

As well as recognising the special rights of those who speak ‘our language’, as outlined above, the Icelandic laws that were in force until the late thirteenth century

(known collectively as Grágás) regularly legislate for the deeds and needs of ‘co- ethnic members’ from other countries. There are provisions for who should prosecute the death of a ‘foreigner from Norway or the realms of the king of

Norway’ who is married in Iceland; if a Dane or Swede or Norwegian should be killed in Iceland, it is anticipated that they might have kin in the country who could prosecute the case; and provision is also made for how to prosecute a killing which took place in one of those countries.156 There are complex inheritance provisions for the eventualities that a man might die abroad, or that his heir might be abroad, with several additions and alterations to the laws, suggesting that there was sufficient to- ing and fro-ing of various kinds to make such changes necessary.157 In some cases it is clear that Norwegians (that is people from the whole realm of the King of Norway, including the ‘colonies’) were granted privileges not granted to other foreigners.158

The Norwegian laws, however, only make rather sparse provision for people from the diaspora, and then only Icelanders, who are to have the same rights to

155 ÍF I, 5. For a more detailed narrative of conflict between Christians and heathens in the earliest stages of the settlement of Iceland, see Landnámabók (Sturlubók chs 23-4, Hauksbók ch 21), ÍF I, 59-65; BoS, pp. 25-6; Jesch 1987. 156 GrágásK I, 170; 172-4; GrágásSt, 387; LEI I, 158, 160, 233. 157 GrágásK I, 226-7, 236-40, II, 195-8; GrágásSt, pp. 70-72, 74-5, 77-9, 88, 95-6, 98-9, 101; LEI II, 8-9, 17-20, 210-14, 244-7. For discussion see Boulhosa 2005, 75-9. 158 GrágásK I, 226; LEI II, 8; Boulhosa 2005, 79.

48 atonement for personal injury as the Norwegian class of men known as hauldar when they are in Norway on trading voyages (though the law also envisages the possibility that they might stay for more than three years, when different provisions apply).159 There is also a provision, presumably applying to both Norwegians and

Icelanders, that, if someone dies in Iceland, or at any point more than halfway between Norway and Iceland, then Icelandic inheritance laws apply.160

After Iceland had lost its independence to Norway, new laws were introduced, based on Norwegian laws but revised for Icelandic conditions. The text known as

Jónsbók was adopted in Iceland in 1281, with parts of it still valid today.161 This law code makes quite clear distinctions between Norway and the Norwegians on the one hand, and Iceland and the Icelanders on the other. The Norwegians are often referred to as útlenzkr ‘foreign’, while Iceland is hér ‘here’ and one of its inhabitants várr landi ‘our countryman’.162 Nevertheless, the law still anticipates the possibility that a Norwegian dying in Iceland may have closely related kin in Iceland to take charge of the inheritance, but since the provision envisages the inheritance being taken back to Norway, we may assume any such kin would be travelling with the deceased, perhaps on a merchant ship.163 The provision also extends to and

Danes, but those ‘of all languages other than Norse [danskri tungu]’ can only inherit if they are father, son or brother.

159 NGL I, 71; ENL, 145; Boulhosa 2005, 82-3. See also LEI II, 210, n. 96, and Boulhosa 2005, 61, on the ‘Norwegian perspective’ of some of the Icelandic provisions. 160 NGL I, 210; ENL, 332. 161 Jb, pp. xi-xiii. 162 E.g. Jb, pp. 30-31, 124-5. 163 Jb, pp. 124-5.

49

The earlier Icelandic laws of Grágás on the other hand make provision for people from quite a wide range of countries, most of which can be considered to be a part of the Viking diaspora, though some other countries where the links might be more of just a trading kind are also mentioned. Greenland and sometimes Norway and the Scottish Isles are presented as places where the law is essentially the same as in Iceland, except that there are some matters that can only be dealt with in one place or the other and, in the case of Norway, some differences are emphasised.164

The laws even make provision for getting rid of an unwanted illegitimate child fathered by someone from Norway and its wider realm, including Shetland, Orkney,

Faroe and Caithness, in which case the child can be handed over to someone from the same region as the father.165 The equivalent provision in Jónsbók implies a rather different social and economic context.166 Firstly, it omits Caithness, more importantly, there is no provision for the child to be sent to its father. Rather, the law requires the father to leave money behind for it and, if he does not, then the law stipulates that a letter is to be written í þann kaupstað er sá maðr sigldi af ‘to that market town where that man sailed from’, requiring various local officials to send the money. It seems that the kinds of situation which could give rise to such problems were by this period most likely to be in a trading context.

The earlier laws also present a wide picture of the places Icelanders might have been expected to go in Britain and Europe. One particularly complex provision in Grágás envisages who can vouch for the circumstances of killings that took place either in ‘western Europe north of Valland [i.e. north of and the lower

164 GrágásK I, 205, 239-40, 249, II, 195-7; GrágásSt, 389; LEI I, 184, 235; II, 19-20, 27, 210-13. 165 GrágásK II, 25; LEI II, 49. 166 Jb, pp. 140-41.

50

Seine region]’ or ‘south of the realm of the Danes’, the guarantors being men who had been ‘in the realm of the king of the English or the king of the Welsh or of the king of the Scots or the king of the Irish or the king of the Hebrideans’.167

‘The possibility of a distinctive creative, enriching life in tolerant host countries’

Some diaspora theorists have emphasised the cultural consequences of diaspora, its enhancement of creativity and cultural production.168 The most obvious example here is the literature of medieval Iceland. This literary production is very considerable in both quantity and chronological extent and can by no means entirely be attributed to diasporic processes. However, it is noticeable that in the twelfth and thirteenth centuries in particular there is extensive literary interchange between

Iceland, Norway and Orkney, much of this based on cultural and historical traditions common to all three areas, which seems to have acted as a spur to some of the later literary developments in Iceland.169 For example, the poem Háttalykill, conceived both as a key to metres and a chronological history of Scandinavian rulers, was composed in mid-twelfth-century Orkney in a collaboration between an Icelandic poet Hallr Þórarinsson and the Orcadian earl Rǫgnvaldr Kali Kolsson.170 It not only provided a model for Snorri Sturluson’s key to metres, the Háttatal section of his

Edda, but is also an early example of the impulse to compose native history, which seems to have started in poetical form before the flowering of historical narratives in prose from the end of the twelfth century. The geographical and historical scope of

167 GrágásSt, pp. 388-9; LEI I, 234. 168 E.g. Kalra et al. 2005, 13, 37. 169 Jesch 2009b. 170 ÍF XXXIV, 185; OS, p. 269; SPSMA III, forthcoming.

51

Háttalykill also prefigured the Norwegian and Icelandic writing of Scandinavian history from the mid-twelfth century onwards.

In an earlier, pre-literate period, rulers of both Scandinavian and native origin in England, Scotland and Ireland employed poets to sing their praises using the very distinctive metres and diction of skaldic poetry.171 This poetry had its origins in early

Viking Age Norway but most, though by no means all, of these praise poets were

Icelanders. Skaldic verse can be found in the Scandinavian homelands in runic inscriptions from the late tenth century to the fourteenth and in both Norwegian and

Icelandic manuscripts of a variety of genres. It was the Icelanders in particular who were able to exploit their expertise in this kind of poetry as a form of cultural capital throughout the diaspora, and this topic will be explored further in Chapter 6.

While the spread of poetry throughout the diaspora was largely a matter of extending certain cultural practices to new audiences and arenas, the Viking Age sculpture of the British Isles is an example of a whole new cultural product that came into being in the creative crucible of the Viking diaspora. With its origins in the ecclesiastical sculpture of early Anglo-Saxon England, combined with various artistic and cultural traditions from the Scandinavian homelands, the sculpture of northern

England in particular is not just an amalgamation of its main sources, but ‘a creative manifestation of religious and cultural integration’ with ‘a strong intellectual component’.172 Many of these monuments are crosses or cross-slabs, and thus clearly derived from insular models. However, there is also one type of monument, the hogback, which has no obvious direct predecessors in either Britain or

171 SPSMA I, cxlix-cli; Abrams 2012, 25. 172 Kopár 2012, xxiv.

52

Scandinavia, and is clearly a new form which arose in an Anglo-Scandinavian context, though it also spread to Scotland, and possibly Wales, and has been described as a ‘Viking colonial monument’.173 The ‘colonial’ aspect of hogbacks may be disputed. Though it is common to view Anglo-Scandinavian sculpture as an assertion of local, secular lordship, this is only one possible explanation and is difficult to demonstrate from the surviving evidence.174 What is clear is that these monuments represent a distinctive creativity that was capable of producing new artistic forms, and the Christian nature of these secular monuments also suggests a relatively easy accommodation of the incomers (presumably after their conversion) with their ‘tolerant host communities’.

Themes of the Viking Diaspora

All of the above demonstrate that the ‘Viking diaspora’ is not simply a new name for the Viking Age migrations, as is sometimes the case in recent usage.175 Rather it is a phenomenon that has its origins in these migrations, but transcends them in time, through linguistic and cultural contacts assiduously maintained throughout the Viking world for some centuries after the migrations. As Kalra et al. put it:

If there is any single theme that emerges from a study of diaspora, it is that of its multi-locational qualities, or the interaction between homes and abroads which cannot be reduced to one place or another.176

173 Lang 1972-4; Lang 1984; Abrams 2012, 36; Kopár 2012, 197-8. 174 Hadley 2006, 221-2; Kopár 2012, 200-202. 175 E.g. Barrett 2010, 289. 176 Kalra et al. 2005, 17-18.

53

‘Diaspora’ thus relates to the processes and results of migration and to how migrants themselves think and feel about their situation. An understanding of these processes requires the following questions to be addressed, though not all of them are easily answered: Did they migrate in a group? Did they migrate to a place where there were other migrants from their home? Did they take their own social and cultural customs with them or did they adopt new ones? Did they give their children traditional names and encourage them to speak the old language as well as the new one? Did they assimilate into the culture of their new homes and if so how many generations did that take? (How indeed does one define ‘assimilation’?) Did they still have any connection with their homeland? If not, did they nevertheless have a sense of where they had come from and a memory of how things were there? And were they in touch with other migrants from the same homeland who had migrated somewhere else entirely? In other words, diaspora is about the migrant’s sense of connectedness:

 to the homeland;  to other migrants from the homeland;  to other regions with migrants from the same homeland;  to their new home.

Once the actual migrations had taken place, there remained webs and networks of connections between all of these groups, and that is what ‘diaspora’ refers to, an ongoing connectedness that came out of a migrational event or events, though with time these webs and networks might become narrower or more specialised.

54

The remaining chapters of this book will look at the implications of a diasporic framework for understanding the following important themes characteristic of the long Viking Age:

 gender and family;

 cults, beliefs and myths;

 networks and identities.

Migration and diaspora are processes which involve individuals, families and communities. The next chapter sharpens this focus by considering the significance of the linguistic, artefactual and natural evidence for understanding gender roles in migration and diaspora.

55